FLUX VERSIONS


Version 0.3-0

  * New functions reco.bulk and gpp.bulk for
  bulk fitting of GPP and Reco models.

  * New functions budget.gpp and budget.reco
  for estimating fluxes with bulk fitted
  models and logger data

  * New function budget.ie for estimating
  interpolation errors introduced by inter-
  polating fluxes across models in a time
  series

  * New helper function tbl8 for getting easy 
  access to model parameters

  * New function modjust that provides some
  ways to globally adjust fitted Reco and GPP 
  models

  * New function gpp2 that offers a new approach
  to GPP modelling: Reco values are directly taken
  from the provided flux data

  * New round functions for time classes POSIXlt
  and POSIXct

  * Bug fixes and consistency changes to gpp, reco
  plot.gpp

  * New function plot.reco

  * General overhaul to the help and examples

  * New data set Ahlenmoor

Version 0.2-2

  * Bug fixes and extensions to gpp. The nls that
  is used internally for fitting the non-linear
  models can get trapped into local minima. Using
  random starts solves that issue and assures that 
  the best model (lowest RSS) is found.

  * Polish to chop. Chopping is now faster for
  big data sets.

  * Addition to mf.flux. Flux estimation can now
  be done consecutively as well. Further, bug fixes
  to mf.flux. Long concentration measurement series
  could cause an error. Now these are handled 
  correctly. Also, fixed intervals are handled 
  better now.

  * Bug fix (new argument) to auc. Catches possible
  errors in data structure (unordered x) of the 
  input data.

Version 0.2-1

  * Minor bug fixes to mf.flux, fluxx and to the 
  namespace. Added methods for evaluating linearity
  to mf.flux and a shortcut to find the linear part
  under certain cicumstances.

  * New argument to plot.gpp for allowing bulk 
  plotting

  * Examples added to the help for mf.flux, fluxx, 
  gpp, reco, and plot.gpp

  * New example data for use in help: tt.nee and 
  tt.flux


Version 0.2-0

	* Housekeeping and bug fixes to many functions.
	
	* New functions mf.flux and fluxx for estimating 
	(ghg) flux rates from online dynamic closed 
	chamber measurements in through-flow mode. Also
	new are corresponding print and plotting methods.
	
	* New functions reco and gpp for modelling NEE
	
	* Function conz.prep has been renamed to chop
	and cleaned up to be more efficient. The argument
	column has been removed.
	
	* New function auc for integrating the area under
	a curve. Was moved to package flux from package 
	simba. In addition, now a threshold for y-values
	can be chosen freely. Further the density of 
	points is  increased for getting more exact 
	results especially when thresholds are set.
	
	* Bug fixes and changes to flux.calib. The
	function is now faster and simpler to use.
	
	* Changed gflux. Now it can also take delta 
	concentration as the first argument. Allowed
	streamlining within the flux functions.
	
	* Bug fixes in flux. Leak flags are now handed 
	over correctly to the flux results list and 
	appear correctly in diagnostic figures.
	
	* Changes to flux and flux.odae that allow to
	hand through factors as well. Before you could
	hand them through but they came out as numerals.


Version 0.1-6

	* Bug fix to conz.prep. Before certain date 
	formats (including "." as separators) could cause
	problems.
	
	* plot.fluxes now warns when the number of
	figures does not match the number of plot
	positions on the sheets.
	
	* Update to flux.odae. The concentration 
	measurements are now ordered according to the
	closure time. Just in case the data are provided
	in wrong order. The optimization procedure did 
	this always but when plotting the correct order 
	is necessary.
	

Version 0.1-5

	* The functionality of flux() and fluxes() 
	is now in flux() including pretty printing
	to the console. fluxes() has been abandoned.
	
	* The range limits are now returned with the flux
	results
	
	* New export() function for simple export.
	
	* The behaviour of inspect() has been changed.
	The function now understands lazy naming: So only 
	the parts of the measurement names that are 
	necessary for unambiguous identification have to
	be given. Therefore, the sustain-argument has 
	been abandoned.
	
	* New plot.fluxes() function. Additionaly the 
	maximum range of xlim in a dataset is taken per 
	default for the xlims of the single figures to 
	ease comparison.
	
	* When prettily printing to the console the
	names of the gases are now printed for objects
	of class fluxes.
	
	* Also the leak.flag can be hardflagged now.
	
	* The gc quality flags can now be missing. In this
	case the function simply assumes that all 
	measurements were OK.
	
	* Function CO2.control() has been abandoned. 
	Because the new fluxes() handles co2ntrol in
	a different way than flux() did it is not necessary 
	anymore.
	
	

Version 0.1-4

	* Considerable changes to almost all functions
	including code clean-up and discontinuation
	of some arguments (e.g. rl.backup and in.unit 
	in flux, start in flux.odae).	
	
	* New function fluxes for estimating fluxes for
	all three major greenhouse gases at once.
	
	* New function inspect for simple inspection
	and alteration of concentration measurement tables.
	
	* New print functions for objects of classes fluss
	and fluxes to give comprehensive but easy to grasp
	output to the console
	
	* Update to flux: argument column has been changed 
	to var.par and allows better control (i.e.
	the input of variables and fixed parameters).
	This also allows to hand through data that 
	shall later be used in further analyses.
	
	* Further updates to flux: Option for p values 
	for the fitted models in the result, 
	
	* Air pressure still defaults to 101325 Pa in 
	the low level function gflux but must be also
	given in flux and related functions through 
	var.par as a fixed parameter or a variable
	
	* Output units are per default determined auto-
	matically to give pretty numbers (0.01 < x <= 10).
	The possibility 
	
	* Flags are global and appear in the same way in
	the table outputs and figures. Via the new 
	argument hardflag flags can be changed from soft
	to hard flag. Per default only the range flag is
	a hard flag (matches the former behaviour).
	
	* Update to CO2.control: max.nrmse can be specified
	separately.
	

Version 0.1-3
	
	* Bug fixes to flux.calib and flux
	* Added argument to function flux (co2ntrol).
	* New function CO2.control that controls how
	the evaluation of the CO2 data influences the 
	flux rate estimation for the other gases
	* New print.fluss function. The results of the
	flux() estimation are now displayed nicely as
	a table and the complex result is hidden.


Version 0.1-2
	
	* Several minor fixes to flux-package.Rd
	* Bugfix to plot.fluss.


Version 0.1-1
	
	* The first version. Could still be unstable.